Skip to content

Latest commit

 

History

History
156 lines (103 loc) · 5.13 KB

README.md

File metadata and controls

156 lines (103 loc) · 5.13 KB

esm-seedrandom

Explicitly seeded random number generator for JavaScript, ported to ES Modules.

Unit tested for number generator compatability with original seedrandom CommonJS NPM package.

  • Version: 3.0.5
  • Author: David Bau
  • Date: 2019-09-14
  • ES Modules port in 2020-12 by Shane Holloway

Demo

Live demo

Use

Fast PRNG Algorithms

PRNG name Period Author BigCrush test results
prng_alea ~2^116 Baagøe pass all
prng_xor128 2^128-1 Marsaglia fail MatrixRank and LinearComp
prng_tychei ~2^127 Neves/Araujo (ChaCha) pass all
prng_xorwow 2^192-2^32 Marsaglia fail CollisionOver, SimpPoker, and LinearComp
prng_xor4096 2^4096-2^32 Brent (xorgens) pass all
prng_xorshift7 2^256-1 Panneton/L'ecuyer pass all
prng_arc4 ~2^1600 Bau (ARC4) unknown

To use Johannes Baagøe's extremely fast Alea PRNG:

// Use alea for Johannes Baagøe's clever and fast floating-point RNG.
import {prng_alea} from 'esm-seedrandom';
let myrng = prng_alea('hello.');

// By default provides 32 bits of randomness in a float.
console.log(myrng());               // Always 0.4783254903741181 for this seed and sequence
console.log(myrng.quick());         // Always 0.8297006865032017 for this seed and sequence

// Use "double" to get 56 bits of randomness.
console.log(myrng.double());        // Always 0.4692433053279662 for this seed and sequence

// Use "int32" to get a 32 bit (signed) integer.
console.log(myrng.int32());         // Always 1350551666 for this seed and sequence

or direclty from HTML,

<script type="module">
  import {prng_alea} from '//cdn.jsdelivr.net/npm/esm-seedrandom/esm/alea.min.js'

  let myrng = prng_alea('an example seed string')

  console.log(myrng()); // Always 0.2594452982302755 for this seed and sequence
  console.log(myrng()); // Always 0.8253263409715146 for this seed and sequence
  console.log(myrng()); // Always 0.42280301195569336 for this seed and sequence

  // Use "quick" to get only 32 bits of randomness in a float.
  console.log(myrng.quick()); // Always 0.9045045920647681 for this seed and sequence
  console.log(myrng.quick()); // Always 0.7626296668313444 for this seed and sequence

  // Use "int32" to get a 32 bit (signed) integer
  console.log(myrng.int32()); // Always 1157605039 for this seed and sequence
  console.log(myrng.int32()); // Always 346379077 for this seed and sequence

  console.log(myrng.double()); // Always 0.9541419381134651 for this seed and sequence
  console.log(myrng.double()); // Always 0.7982540860513401 for this seed and sequence
</script>

Docs

See the API docs.

Overview

From NodeJS,

npm install esm-seedrandom

or in HTML,

<script type='module'>
  import {prng_alea} from '//cdn.jsdelivr.net/npm/esm-seedrandom/esm/index.min.js'

  // or use the individual algorithms by module

  import {prng_alea} from '//cdn.jsdelivr.net/npm/esm-seedrandom/esm/alea.min.js'
  import {prng_xor128} from '//cdn.jsdelivr.net/npm/esm-seedrandom/esm/xor128.min.js'
  import {prng_tychei} from '//cdn.jsdelivr.net/npm/esm-seedrandom/esm/tychei.min.js'
  import {prng_xorwow} from '//cdn.jsdelivr.net/npm/esm-seedrandom/esm/xorwow.min.js'
  import {prng_xor4096} from '//cdn.jsdelivr.net/npm/esm-seedrandom/esm/xor4096.min.js'
  import {prng_xorshift7} from '//cdn.jsdelivr.net/npm/esm-seedrandom/esm/xorshift7.min.js'
  import {prng_arc4} from '//cdn.jsdelivr.net/npm/esm-seedrandom/esm/arc4.min.js'
</script>

Saving & Restoring PRNG state

import {prng_alea} from 'esm-seedrandom';

let rng_first = prng_alea("secret-seed", {state: true});
let saved_state = rng_first.state()
for (let j = 0; j < 1e5; ++j)
  rng_first();

// later

let rng_replica = prng_alea("", {state: saved_state});
for (let j = 0; j < 1e5; ++j)
  rng_replica();

In normal use the prng is opaque and its internal state cannot be accessed. However, if the state option is provided, the prng gets a state() method that returns a plain object the can be used to reconstruct a prng later in the same state (by passing that saved object back as the state option).

Unit tests

Mocha-based live unittests for reproducability and validation of state capture and restore.

In NodeJS unittests, validation of state snapshot compatability with the original seedrandom CommonJS implementation is performed.

License (MIT)

MIT